Distributed and accumulated reinforcement arrangements: evaluations of efficacy and preference.

de Kuijper et al. (2014) asked a simple question: Is it better to give tiny bits of reinforcer after each response or let learners save up and cash in later?

They used an alternating-treatments design with adults who had intellectual disabilities. Each person tried both ways in mixed order during the same session.

Banking the reinforcer and taking it all at once kept response rates just as high as little payouts. Most participants actually liked the saved-up version better, even though they had to wait.

Weston et al. (2020) ran the same comparison with autistic children using token boards and got the same lift in work rate with saved-up rewards. The pattern holds across diagnoses and setups.

Frank-Crawford et al. (2021) and Fulton et al. (2020) moved the idea into problem-behavior treatment. Both teams found that big delayed breaks cut escape-maintained behavior at least as well as, and sometimes better than, tiny frequent ones.

Chen et al. (2022) tested feeding disorders and saw a twist: some kids still preferred bite-by-bite praise. So the saved-up plan is not a universal favorite; you still need to ask each learner.

You can stop worrying that delayed reinforcement will weaken performance. Letting clients earn a single 5-minute cartoon or snack pile at the end of work keeps responding strong and often boosts satisfaction. Start by offering both options during a probe, note which one the learner picks, and run with it. You may free yourself from constant resetting of 30-second timers and give richer, more natural reinforcement moments.